Sunday, August 31, 2014
Big Data in Forecasting
After reading all the articles related to this week’s theme, I would like to discuss the issue of forecasting in retail industry. As it is a low-margin business compared with other kinds of industries, the analytics of customers’ preferences, including what and where they are going to purchase, which kind of promotion will actually enhance the buying process, will make a great difference and define the winners and losers in the retail industry.
As a matter of fact, more and more retail companies are utilizing big data to analyze and forecast. For instance, Macy’s Inc. “is using analytic software from SAS to better understand and enhance its customers’ online shopping experience, while helping to increase the retailer’s overall profitability.” By using SAS system Macy’s create the real-time pricing system to regulate the prices of goods based on the demand of customers and the inventory information. This is a typical model of good practice in forecasting as Macy’s take advantage of good forecasting software to make proper real-time decisions.
However, the information era is so all-pervasive that every individual is deluged with tons of information, which makes customers are now informed with so many deals gathered from various kinds of channels and they can interact and talk with friends concerning shopping locations and items of their wish lists. Besides, people can choose to shop in brick- and-mortar stores or simply shop online without traveling outside. The multichannel choices benefit customers while bring difficulties to retailers in keeping the in-stock status, which is hard to sustain when both physical and digital shopping proportions in different goods can not be simply perceived. In this circumstance, it is more complicated for retailers to forecast, which pushes them to gather more information from more data sources so as to target the precise customers that have more potential to finish the buying process. The retailers are supposed to gather information not only concerning online and physical stores’s sales and inventory levels but also other data when it comes to big promotions. For example, A 2013 report, “The Emerging Big Returns From Big Data”, mentioned a data source, that is mobile phone information obtained from telecom operators could alert retailers to when customers are within five miles of a store. All these capturing and using of sources and big data are excellent practices of forecasting.
As far as I am concerned, the forecasting and analytics are crucial for retail industry as long as they are eager to be the winners in the retailing landscape and all the examples above are good models for other retails to follow. Nevertheless, practical ways to do efficient forecasting are being excavated especially for those retail companies with razor-thin profits.